"What is interesting about perovskite is that all the research groups — in Korea, England, Switzerland, the United States—they're all getting very high efficiencies," Luther said. "It's not as if just one person knows the secret."
The theoretical maximum efficiency of a perovskite-based solar cell is about 31 percent, meaning that of all the solar energy contained in the sunlight that hits the cell, 31% is converted to useful electrical energy. Multijunction cells based on perovskites could attain higher efficiencies still.
"The goal shouldn't be to stop at 20 percent efficiency," Luther said. "The goal should be to try to get to 28% or higher. In the lab, the best cells need to be almost perfect at small scale. Then the commercial people can stop at whatever efficiency is economical for them to deploy."
NREL has world-class experts in several fields needed for the exploration and improvement of this new promising material: experts in so-called III-V cells from the third and fifth columns of the periodic table; in quantum dots, materials, and transport; in computational materials design; and in doping materials with new materials to change their band gaps, and thus their usefulness in harvesting electrons.
In fact, NREL's latest world record has echoes of properties inherent in perovskite. NREL recently set a world record of 34% conversion efficiency for a gallium indium phosphide cell atop a gallium arsenide cell under lenses that multiply the sun's power. Last June, NREL set the world record of 31.1% for the same cell under one sun. In both cases, NREL reached unprecedented efficiencies by improving the ability of electrons to diffuse out of their traps. It's that long-diffusion, short-absorption phenomenon that has scientists so excited about perovskite.
Remarkable Progress in Just Five Years
NREL Senior Scientist Kai Zhu is co-organizer of a scientific conference on dye-sensitized solar cells, which are low-cost thin-film cells. So far, he says, the majority of talks, posters, and papers proposed for the conference are on the subject of perovskite—so exciting is the field even though perovskite isn't technically a dye cell.
Progress in efficiency for solar cells made from perovskite absorbers. Efficiency values were taken from publications and NREL's latest chart on record cell efficiencies.
In 2009, Japanese scientist Tsutomu Miyasaka reported perovskite's potential as a light absorber and possible material for a solar cell, noting a 3.8% conversion efficiency, but that was such a low rate that it didn't spark much interest, said Zhu.
But in 2011, a Korean scientist, Nam-Gyu Park, who served as a postdoctoral researcher at NREL in the late 1990s, reported achieving 6.5% efficiency with perovskite. "Fifteen years ago, he was working in the same lab I am working in now," Zhu said. "So I started paying attention to his work on perovskites."
A year later, Michael Grätzel, a top solar scientist from Switzerland, teamed with Park on a paper, sparking more widespread interest. Their paper in the journal Nature Scientific Reports reported a conversion efficiency of about 10% with perovskite. "By then, I knew this was something I wanted to pursue," Zhu said. At the beginning of 2013, the efficiency level for perovskite had climbed to 12.3%.
"And then about a year ago, when they added chlorine to the materials, the electron and hole diffusion lengths just went through the roof," Ginley said. "The most remarkable thing is that you add a little bit of chlorine and you see how the diffusion lengths change—by a factor of 10. That really brought attention to them." Ideally, a solar cell has a diffusion length long enough for the electron to reach the contacts both above and below it, and thus escape the possibility that it will be trapped in its layer and recombine into an electron-hole pair.
When Zhu's proposal to examine perovskite was approved, the efficiency level had climbed to 14.1%. Now, the highest certified rate is 16.2% by Sang Il Seok of Korea. "Seeing how rapidly this field is progressing, I feel very lucky that I started on this more than a year ago," Zhu said.
Meanwhile, Zhu is in the midst of an experiment in which he prepares a precursor solution that converts from a liquid base to an absorber in a device. "This material is so easy to work with," Zhu said. "Working on solution processing, we can make a device in one or two days, from beginning to finish."
To boost efficiency levels even further will take more effort, Zhu concedes. "But this new material can probably be processed at a much lower cost" than rival materials, he said. It doesn't have to deal with the problem of the substrate not matching with the material above it, or with the delicate deposition process necessary with many alternative solar materials.
Several companies are already interested in forming cooperative research and development agreements so they can work with NREL on perovskite. "At NREL, we have this depth and breadth of understanding of materials, devices, transport, and, really, all aspects of solar cells that should help us make an important contribution to this new material," Zhu said.